WO2021143370A1 - 资源数据的处理方法及装置 - Google Patents

资源数据的处理方法及装置 Download PDF

Info

Publication number
WO2021143370A1
WO2021143370A1 PCT/CN2020/132620 CN2020132620W WO2021143370A1 WO 2021143370 A1 WO2021143370 A1 WO 2021143370A1 CN 2020132620 W CN2020132620 W CN 2020132620W WO 2021143370 A1 WO2021143370 A1 WO 2021143370A1
Authority
WO
WIPO (PCT)
Prior art keywords
resource
data
value
division
target
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2020/132620
Other languages
English (en)
French (fr)
Inventor
倪翔
汲小溪
王维强
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Alipay Hangzhou Information Technology Co Ltd
Original Assignee
Alipay Hangzhou Information Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Alipay Hangzhou Information Technology Co Ltd filed Critical Alipay Hangzhou Information Technology Co Ltd
Publication of WO2021143370A1 publication Critical patent/WO2021143370A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/232—Non-hierarchical techniques
    • G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
    • G06F18/23211—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with adaptive number of clusters
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0635—Risk analysis of enterprise or organisation activities
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
    • G06Q40/04—Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange

Definitions

  • This document relates to the technical field of data processing and risk assessment, in particular to a method and device for processing resource data.
  • the risk management of these merchants includes four main links: merchant access, merchant risk identification, merchant risk operation, and merchant risk decision-makers.
  • risk control operations, strategies, and models are based on various data of merchant entities, but in fact, there are inextricably linked relationships between different merchant entities (such as overlapping identities, fund exchange relationships, etc.) , Media sharing relationship, geographical proximity, etc.), if different subjects can be aggregated based on the connections between subjects, the data of different subjects can be opened, and the analysis, control and modeling of each subject can be effectively Improve the accuracy, coverage and effectiveness of risk management to achieve joint ecological risk prevention.
  • quantiles An important link in aggregate calculation is the calculation of quantiles.
  • the calculation of quantiles is very common in descriptive statistics. For example, compared with the average, the median will not be affected by outliers, but the calculation process of quantiles is more complicated, and all specific values need to be kept. After sorting Get the number in the middle position as the result. This complex quantile calculation method leads to a decrease in the efficiency of the main body aggregation and a decrease in the efficiency of the main body management (such as main body risk prevention and control).
  • one or more embodiments of the present specification provide a resource data processing method, including: determining at least one resource division value corresponding to the multiple resource data based on the division positions of the multiple resource data.
  • the first degree of difference between the resource division value and the extreme value of the resource data is smaller than the second degree of difference between the division position and the extreme value.
  • the extreme value includes the maximum data value and/or the minimum data value in the resource data.
  • a target resource cluster group corresponding to the resource evaluation index is determined from the multiple resource cluster groups.
  • the resource evaluation index includes risk evaluation parameters used for risk evaluation of the resource data.
  • the resource assessment threshold is used to perform risk assessment on the resource data.
  • one or more embodiments of the present specification provide a resource data processing device, including: a first determining module, which determines at least one resource partition corresponding to the multiple resource data based on the division positions of the multiple resource data value.
  • the first degree of difference between the resource division value and the extreme value of the resource data is smaller than the second degree of difference between the division position and the extreme value.
  • the extreme value includes the maximum data value and/or the minimum data value in the resource data.
  • the clustering module uses each of the resource division values to perform clustering processing on the multiple resource data to obtain multiple resource clustering groups.
  • the second determining module determines the target resource cluster group corresponding to the resource evaluation index from the plurality of resource cluster groups according to the preset resource evaluation index.
  • the resource evaluation index includes risk evaluation parameters used for risk evaluation of the resource data.
  • the third determining module determines the resource evaluation threshold corresponding to the resource data according to the target resource segmentation value corresponding to the target resource clustering group.
  • the resource assessment threshold is used to perform risk assessment on the resource data.
  • one or more embodiments of the present specification provide a resource data processing device, including: a processor; and a memory arranged to store computer-executable instructions that, when executed, cause the Processor: Based on the division positions of the multiple resource data, determine at least one resource division value corresponding to the multiple resource data. The first degree of difference between the resource division value and the extreme value of the resource data is smaller than the second degree of difference between the division position and the extreme value. The extreme value includes the maximum data value and/or the minimum data value in the resource data. Perform clustering processing on the multiple resource data using each of the resource division values to obtain multiple resource clustering groups. According to a preset resource evaluation index, a target resource cluster group corresponding to the resource evaluation index is determined from the multiple resource cluster groups. The resource evaluation index includes risk evaluation parameters used for risk evaluation of the resource data. Determine the resource evaluation threshold corresponding to the resource data according to the target resource segmentation value corresponding to the target resource clustering group. The resource assessment threshold is used to perform risk assessment on the resource data.
  • one or more embodiments of this specification provide a storage medium for storing computer-executable instructions that, when executed, realize the following process: based on the division positions of multiple resource data, determine all At least one resource division value corresponding to the plurality of resource data.
  • the first degree of difference between the resource division value and the extreme value of the resource data is smaller than the second degree of difference between the division position and the extreme value.
  • the extreme value includes the maximum data value and/or the minimum data value in the resource data.
  • a target resource cluster group corresponding to the resource evaluation index is determined from the multiple resource cluster groups.
  • the resource evaluation index includes risk evaluation parameters used for risk evaluation of the resource data. Determine the resource evaluation threshold corresponding to the resource data according to the target resource segmentation value corresponding to the target resource clustering group.
  • the resource assessment threshold is used to perform risk assessment on the resource data.
  • Fig. 1 is a schematic flowchart of a method for processing resource data according to an embodiment of the present specification
  • Fig. 2 is a schematic flowchart of a method for processing resource data according to another embodiment of the present specification
  • Fig. 3 is a schematic block diagram of an apparatus for processing resource data according to an embodiment of the present specification
  • Fig. 4 is a schematic block diagram of a resource data processing device according to an embodiment of the present specification.
  • One or more embodiments of this specification provide a method and device for processing resource data to solve the problems of low data clustering efficiency and low risk management efficiency.
  • Fig. 1 is a schematic flowchart of a method for processing resource data according to an embodiment of the present specification. As shown in Fig. 1, the method includes steps S102 to S108.
  • S102 Determine at least one resource division value corresponding to the multiple resource data based on the division positions of the multiple resource data.
  • the first degree of difference between the resource division value and the extreme value of the resource data is smaller than the second degree of difference between the division position and the extreme value.
  • the extreme value includes the maximum data value and/or the minimum data value in the resource data. For example, if the resource data is arranged in the order of data size, then the extreme values are the top resource data and the last resource data.
  • the resource division value is generally between 0 and 1, which indicates the division position of the resource data. For example, if the resource division value is 20%, it means that the resource data is divided at 20%.
  • the first degree of difference between the resource segmentation value and the extreme value of the resource data is the closeness of the resource segmentation value to the extreme value of the resource data.
  • the second degree of difference between the extreme values of the segmentation data and the resource data is the closeness of the extreme values of the segmentation data and the resource data.
  • S104 Perform clustering processing on multiple resource data using each resource segmentation value to obtain multiple resource clustering groups.
  • S106 According to a preset resource evaluation index, determine a target resource cluster group corresponding to the resource evaluation index from a plurality of resource cluster groups.
  • resource evaluation indicators include risk evaluation parameters used for risk evaluation of resource data.
  • Risk assessment parameters such as the proportion of low-risk data in all resource data. For example, if the risk assessment parameter is that low-risk data accounts for 90% of all resource data, it means that the preset resource assessment index hopes to perform risk assessment on resource data when the proportion of low-risk data is 90%.
  • other resource evaluation indicators can also be preset, such as the total number of low-risk data or the total number of high-risk data.
  • S108 Determine a resource assessment threshold corresponding to the resource data according to the target resource segmentation value corresponding to the target resource clustering group, where the resource assessment threshold is used for risk assessment of the resource data.
  • the resource assessment threshold can be determined based on the preset resource assessment index, which not only improves the accuracy of the resource assessment threshold, but also makes the risk assessment result of the resource data more accurate.
  • the Tdigest algorithm can be used to cluster the resource data.
  • TDigest is a simple, fast, accurate, and parallelizable approximate percentile algorithm.
  • the core idea of the TDigest algorithm is Sketch, which is an abstract and simplified idea and a way to deal with problems. The following details how to use the Tdigest algorithm to cluster resource data.
  • At least one division position for dividing multiple resource data may be determined first; and then according to the mapping relationship between each division position and resource division value, the resource division corresponding to each division position is calculated. value.
  • the division position may be in the form of a percentage
  • the resource division value corresponding to the division position may be the centroid of the resource data.
  • the mapping relationship between the segmentation position and the resource segmentation value (that is, the mapping relationship between the percentage and the centroid) is as shown in the following formula (1).
  • q represents the percentage
  • k represents the centroid
  • ⁇ is a constant, usually a relatively small value, such as 0.01. The value of ⁇ affects the size of the centroid k.
  • the division position can be converted into the resource division value through the mapping relationship between each division position and the resource division value. Since the difference between the resource division value and the extreme value of the resource data is smaller, it can make The clustering results of resource data are more accurate and faster.
  • each resource segmentation value may be sorted according to a preset dimension, and the preset dimension includes the data size of the resource data. Then, based on the sorted resource segmentation values, every two adjacent resource segmentation values are determined as the boundary value corresponding to a resource clustering group. Based on the respective boundary values of each resource clustering group, multiple resource clustering groups are determined, and each resource data is divided into corresponding resource clustering groups.
  • the resource division values are sorted according to the size of the resource data, that is, the centroids are sorted according to the size of the centroid. Then determine every two adjacent centroids as the boundary value corresponding to a resource cluster group, and then determine multiple resource cluster groups based on the boundary value corresponding to each resource cluster group, and then divide the resource data separately To the corresponding resource clustering group.
  • each resource data can be sorted according to the data size, and then each resource data can be compared with the corresponding boundary value of each resource clustering group to determine Find out which resource clustering group each resource data falls into, and divide each resource data into the resource clustering group it falls into.
  • the resource data is clustered according to the resource segmentation value (that is, the center of mass), and the resource segmentation value is closer to the extreme value of the resource data, the clustering result is more accurate and more in line with people’s perception of extreme location resources in practical applications. Data care needs.
  • the risk assessment parameter may be the proportion of high-risk data in all resource data. Based on this, if the risk assessment parameter includes the first resource segmentation value, the first resource segmentation value is used to characterize the proportion of high-risk data in all resource data.
  • the resource cluster group where the first resource segmentation value is located can be determined from the multiple resource cluster groups, and the resource cluster group where the first resource segmentation value is located is determined as the target resource cluster group.
  • the weight corresponding to each target resource segmentation value when determining the resource evaluation threshold corresponding to the resource data according to the target resource segmentation value corresponding to the target resource clustering group, may be determined first, and then the weights corresponding to the target resource segmentation values and the respective target resource segmentation values are determined. The weights corresponding to the target resource segmentation values are calculated, and the resource evaluation threshold corresponding to the resource data is calculated. Among them, the sum of the weights corresponding to each target resource segmentation value is 1.
  • the target resource segmentation value corresponding to the target resource clustering group is the two boundary values of the target resource clustering group.
  • the first resource segmentation value is used to characterize the proportion of high-risk data in all resource data.
  • the target resource clustering group includes a first target resource segmentation value and a second target resource segmentation value (that is, two boundary values of the target resource clustering group).
  • the weights corresponding to the respective target resource division values can be determined in the following manner: First, the resource percentile corresponding to the first resource division value is determined. Second, determine the resource percentile corresponding to the first resource segmentation value as the weight corresponding to the first target resource segmentation value. Third, the absolute value of the difference between the resource percentile and 1 is calculated, and the absolute value is determined to be the weight corresponding to the second target resource segmentation value.
  • the target resource cluster group corresponding to the resource evaluation index is determined from a plurality of resource cluster groups according to the preset resource evaluation index, and the resource data corresponding to the resource data is determined according to the target resource segmentation value corresponding to the target resource cluster group.
  • the resource evaluation threshold of enables the resource evaluation threshold to be determined based on the preset resource evaluation index, which not only improves the accuracy of the resource evaluation threshold, but also makes the risk evaluation result of the resource data more accurate.
  • the resource data processing method provided in the foregoing embodiment can be applied to various resource data risk assessment scenarios. For example, risk assessment of transaction amount, risk assessment of business data, etc.
  • the following takes the risk assessment scenario of the transaction amount as an example.
  • Fig. 2 is a schematic flowchart of a method for processing resource data according to another embodiment of the present specification.
  • the resource data is the amount of multiple transactions in a certain period of time.
  • the method includes steps S201 to S208.
  • the percentage is the division position of multiple transaction amounts. For example, it is determined that the percentages for dividing the multiple transaction amounts are 20%, 40%, 60%, and 80%.
  • the determination of the percentage can be specified by the user, or can be determined by the computer according to preset rules.
  • the preset rule may be any of the following rules: randomly determine the percentage from 0 to 1, determine the percentage evenly from 0 to 1 according to a preset interval, determine evenly N percentages from 0 to 1, and so on.
  • S202 Determine multiple amount split values corresponding to multiple transaction amounts based on the multiple determined percentages.
  • the amount split value is equivalent to the centroid of multiple transaction amounts. Knowing the percentages, the above formula (1) can be used to determine the amount split value corresponding to each percentage.
  • the amount division value determined in this step is generally between 0 and 1, which represents the division position of the transaction amount. For example, if the amount split value is 20%, it means that the split is performed at 20% of the transaction amount.
  • the centroid has the following characteristics: the first degree of difference between the centroid and the extreme value of the transaction amount is smaller than the second degree of difference between the percentage determined in S201 and the extreme value of the transaction amount.
  • the extreme value of the transaction amount includes the maximum value and the minimum value of all transaction amounts. That is, converting the percentage into a centroid to divide the transaction amount can make the position of the transaction amount split closer to the extreme value of the transaction amount.
  • S203 Sort the multiple amount division values according to the size of the amount, and use every two adjacent amount division values as the boundary value corresponding to one amount cluster group to obtain multiple amount cluster groups.
  • each transaction amount can be sorted according to the amount of money, and then see which amount cluster group each transaction amount falls into, and then divide each transaction amount into its own amount cluster group.
  • the transaction amount evaluation index includes the risk evaluation parameters used in the risk evaluation of the transaction amount, for example, the proportion of the low-risk transaction amount in all transaction amounts.
  • S206 Determine the target amount cluster group corresponding to the transaction amount evaluation index from the multiple amount cluster groups.
  • the transaction amount evaluation index is 90% of the low-risk transaction amount in all transaction amounts. It is also assumed that the multiple amount split values determined in S202 are respectively: 0.2, 0.8, and 0.95. Then it can be determined that the target amount cluster group corresponding to the transaction amount evaluation index is an amount cluster group composed of the amount split values 0.8 and 0.95, that is, the amount split value 0.8 and 0.95 are used as the amount cluster group of the two boundary values.
  • the specific method of using interpolation to calculate the amount evaluation threshold is as follows: First, determine the percentile of the amount (that is, the transaction amount evaluation index). Second, determine the percentile of the amount as the weight corresponding to the first target amount split value (that is, the smaller target amount split value corresponding to the target amount cluster group), and calculate the difference between the percentile of the amount and 1 The absolute value of the value is determined as the weight corresponding to the second target amount split value (that is, the larger target amount split value corresponding to the target amount cluster group). Finally, based on the respective weights corresponding to each target amount split value and each target amount split value, the amount evaluation threshold corresponding to the transaction amount is calculated.
  • the first target amount split value is a, and its corresponding weight is 0.9; the second target amount split value is b, and its corresponding weight is 0.1.
  • the amount evaluation threshold is: a*0.9+b*0.1.
  • S208 Perform risk assessment on the transaction amount based on the amount assessment threshold.
  • the transaction amount is greater than the amount evaluation threshold, it can be determined that the transaction amount is a high-risk transaction amount; if the transaction amount is less than or equal to the amount evaluation threshold, it can be determined that the transaction amount is a low-risk transaction amount.
  • a plurality of transaction amounts can be clustered based on each amount division value with a smaller degree of difference between the extreme value of the transaction amount, and a plurality of amount cluster groups can be obtained.
  • people are more concerned about the risk of the transaction amount at the extreme position, so the clustering result of the transaction amount can be made more accurate and faster.
  • the evaluation threshold allows the amount evaluation threshold to be determined based on the preset transaction amount evaluation index, which not only improves the accuracy of the transaction amount evaluation threshold, but also makes the risk assessment result of the transaction amount more accurate.
  • one or more embodiments of this specification also provide a resource data processing device.
  • Fig. 3 is a schematic block diagram of an apparatus for processing resource data according to an embodiment of the present specification.
  • the device for processing resource data includes: a first determining module 310, a clustering module 320, a determining module 330, and a third determining module 340.
  • the first determining module 310 determines at least one resource segmentation value corresponding to the multiple resource data based on the segmentation positions of the multiple resource data; the first degree of difference between the resource segmentation value and the extreme value of the resource data is smaller than the segmentation location and the extreme value
  • the second degree of difference between; extreme values include the maximum data value and/or the minimum data value in the resource data
  • the clustering module 320 uses each resource segmentation value to perform clustering processing on multiple resource data to obtain multiple resource clustering groups;
  • the second determination module 330 determines the target resource cluster group corresponding to the resource evaluation index from a plurality of resource cluster groups according to the preset resource evaluation index; the resource evaluation index includes the risk evaluation parameters used for risk evaluation of the resource data ;
  • the third determining module 340 determines the resource assessment threshold corresponding to the resource data according to the target resource segmentation value corresponding to the target resource clustering group; the resource assessment threshold is used for risk assessment of the resource data.
  • the first determining module 310 includes: a first determining unit, which determines at least one segmentation location for segmenting multiple resource data; a calculation unit, which calculates according to the mapping relationship between each segmentation location and the resource segmentation value The resource division value corresponding to each division position.
  • the clustering module 320 includes: a sorting unit, which sorts each resource segmentation value according to a preset dimension; the preset dimension includes the data size of the resource data; and a second determining unit is based on the sorted resource segmentation value , Determining every two adjacent resource segmentation values as the boundary value corresponding to a resource clustering group; the third determining unit determines a plurality of resource clustering groups based on the boundary value corresponding to each resource clustering group;
  • the dividing unit divides each resource data into corresponding resource cluster groups.
  • the third determining module 340 includes: a fourth determining unit that determines the weights corresponding to the target resource segmentation values; wherein the sum of the weights corresponding to the target resource segmentation values is 1; The target resource segmentation value and the weight corresponding to each target resource segmentation value are calculated, and the resource evaluation threshold corresponding to the resource data is calculated.
  • the risk assessment parameter includes a first resource segmentation value; the target resource cluster group includes a first target resource segmentation value and a second target resource segmentation value.
  • the fourth determining unit is further configured to: determine the resource percentile corresponding to the first resource segmentation value; determine the resource percentile corresponding to the first resource segmentation value as the weight corresponding to the first target resource segmentation value; and calculate the resource percentage The absolute value of the difference between the number of bits and 1; the absolute value is determined to be the weight corresponding to the second target resource segmentation value.
  • the second determining module 330 includes: a fifth determining unit, which determines the resource cluster group in which the first resource segmentation value is located from the plurality of resource cluster groups; and the sixth determining unit, which determines the first resource segmentation value The resource clustering group in which it is located is determined as the target resource clustering group.
  • the device of one or more embodiments of this specification it is possible to determine at least one resource segmentation value corresponding to multiple resource data based on the segmentation position of the multiple resource data, and use each resource segmentation value to perform clustering processing on the multiple resource data, Obtain multiple resource clustering groups. It can be seen that the division of resource data is not based on the original division position, but is divided by the calculated resource division value with a smaller difference between the extreme value of the resource data, because in practical applications, people are more Care about resource data in extreme positions, so the clustering results of resource data can be made more accurate and faster.
  • the resource assessment threshold can be determined based on the preset resource assessment index, which not only improves the accuracy of the resource assessment threshold, but also makes the risk assessment result of the resource data more accurate.
  • Resource data processing devices may have relatively large differences due to different configurations or performances, and may include one or more processors 401 and a memory 402, and the memory 402 may store one or more storage applications or data. Among them, the memory 402 may be short-term storage or persistent storage.
  • the application program stored in the memory 402 may include one or more modules (not shown in the figure), and each module may include a series of computer-executable instructions in a device for processing resource data.
  • the processor 401 may be configured to communicate with the memory 402, and execute a series of computer-executable instructions in the memory 402 on the resource data processing device.
  • the resource data processing equipment may also include one or more power supplies 403, one or more wired or wireless network interfaces 404, one or more input and output interfaces 405, and one or more keyboards 406.
  • the resource data processing device includes a memory and one or more programs, wherein one or more programs are stored in the memory, and the one or more programs may include one or more modules, and Each module may include a series of computer-executable instructions in a processing device for resource data, and the one or more programs configured to be executed by one or more processors include computer-executable instructions for performing the following: To determine at least one resource division value corresponding to the plurality of resource data; the first difference between the resource division value and the extreme value of the resource data is smaller than the division position and the The second degree of difference between extreme values; the extreme values include the maximum data value and/or the minimum data value in the resource data; clustering the multiple resource data by using each of the resource segmentation values, Obtain a plurality of resource clustering groups; determine the target resource clustering group corresponding to the resource evaluation index from the plurality of resource clustering groups according to preset resource evaluation indexes; the resource evaluation index includes the resource evaluation index The risk assessment parameters used for risk assessment of data; determine the resource assessment
  • the processor may also cause the processor to: determine at least one of the division positions for dividing the plurality of resource data; and divide the resources according to each of the division positions and the resources. The mapping relationship between the values is calculated, and the resource division value corresponding to each of the division positions is calculated.
  • the processor may also cause the processor to: sort the resource division values according to a preset dimension; the preset dimension includes the data size of the resource data; For each of the resource division values after sorting, each two adjacent resource division values are determined as the boundary value corresponding to a resource cluster group; based on the boundary value corresponding to each resource cluster group, A plurality of said resource clustering groups are determined; each of said resource data is divided into corresponding respective said resource clustering groups.
  • the processor may also cause the processor to: determine the weights corresponding to the target resource segmentation values; wherein the sum of the weights corresponding to the target resource segmentation values is 1. Calculate the resource evaluation threshold corresponding to the resource data according to the respective target resource segmentation value and the weight corresponding to each target resource segmentation value.
  • the risk assessment parameter includes a first resource segmentation value; the target resource clustering group includes a first target resource segmentation value and a second target resource segmentation value; when the computer-executable instruction is executed, it can also use The processor: determining the resource percentile corresponding to the first resource division value; determining the resource percentile corresponding to the first resource division value as the weight corresponding to the first target resource division value; calculating The absolute value of the difference between the resource percentile and 1; determining that the absolute value is the weight corresponding to the second target resource segmentation value.
  • the processor may also cause the processor to: determine the resource cluster group in which the first resource division value is located from among the plurality of resource cluster groups; A resource cluster group where a resource division value is located is determined as the target resource cluster group.
  • One or more embodiments of this specification also propose a computer-readable storage medium that stores one or more programs, and the one or more programs include instructions.
  • the instructions include multiple application programs
  • the electronic device of can execute the above-mentioned resource data processing method, and is specifically used to execute: based on the division position of the plurality of resource data, determine at least one resource division value corresponding to the plurality of resource data; The first degree of difference between the resource division value and the extreme value of the resource data is smaller than the second degree of difference between the division position and the extreme value; the extreme value includes the largest data in the resource data Value and/or minimum data value; clustering the multiple resource data using each of the resource segmentation values to obtain multiple resource clustering groups; according to preset resource evaluation indicators, gathering from the multiple resources
  • the target resource cluster group corresponding to the resource evaluation index is determined in the cluster group; the resource evaluation index includes the risk assessment parameters used for risk assessment of the resource data; the target resource corresponding to the target resource cluster group
  • the segmentation value determines the resource assessment threshold corresponding to the
  • a typical implementation device is a computer.
  • the computer may be, for example, a personal computer, a laptop computer, a cell phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or Any combination of these devices.
  • one or more embodiments of this specification can be provided as a method, a system, or a computer program product. Therefore, one or more embodiments of this specification may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, one or more embodiments of this specification may adopt computer programs implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes. The form of the product.
  • computer-usable storage media including but not limited to disk storage, CD-ROM, optical storage, etc.
  • These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing equipment to work in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including the instruction device.
  • the device implements the functions specified in one process or multiple processes in the flowchart and/or one block or multiple blocks in the block diagram.
  • These computer program instructions can also be loaded on a computer or other programmable data processing equipment, so that a series of operation steps are executed on the computer or other programmable equipment to produce computer-implemented processing, so as to execute on the computer or other programmable equipment.
  • the instructions provide steps for implementing the functions specified in one process or multiple processes in the flowchart and/or one block or multiple blocks in the block diagram.
  • the computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
  • processors CPUs
  • input/output interfaces network interfaces
  • memory volatile and non-volatile memory
  • the memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and/or non-volatile memory, such as read-only memory (ROM) or flash memory (flash RAM).
  • RAM random access memory
  • ROM read-only memory
  • flash RAM flash memory
  • Computer-readable media include permanent and non-permanent, removable and non-removable media, and information storage can be realized by any method or technology.
  • the information can be computer-readable instructions, data structures, program modules, or other data.
  • Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, CD-ROM, digital versatile disc (DVD) or other optical storage, Magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media can be used to store information that can be accessed by computing devices. According to the definition in this article, computer-readable media does not include transitory media, such as modulated data signals and carrier waves.
  • program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types.
  • This application can also be practiced in distributed computing environments. In these distributed computing environments, tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • Physics & Mathematics (AREA)
  • Economics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Development Economics (AREA)
  • Marketing (AREA)
  • Data Mining & Analysis (AREA)
  • General Business, Economics & Management (AREA)
  • Educational Administration (AREA)
  • Game Theory and Decision Science (AREA)
  • Finance (AREA)
  • Accounting & Taxation (AREA)
  • Tourism & Hospitality (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • General Engineering & Computer Science (AREA)
  • Probability & Statistics with Applications (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Technology Law (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

一种资源数据的处理方法及装置，用以解决现有技术中数据聚类效率低以及风险管理效率低的问题。所述方法包括：基于多个资源数据的分割位置，确定所述多个资源数据对应的至少一个资源分割值（S102）。利用各所述资源分割值对所述多个资源数据进行聚类处理，得到多个资源聚类组（S104）。根据预设的资源评估指标，从所述多个资源聚类组中确定所述资源评估指标对应的目标资源聚类组（S106）。所述资源评估指标包含对所述资源数据进行风险评估所使用的风险评估参数。根据所述目标资源聚类组对应的目标资源分割值，确定所述资源数据对应的资源评估阈值，所述资源评估阈值用于对所述资源数据进行风险评估（S108）。

Description

资源数据的处理方法及装置 技术领域
本文件涉及数据处理及风险评估技术领域，尤其涉及一种资源数据的处理方法及装置。
背景技术
在风控场景中涉及到的商户主体主要有三种类型：直连商户、间连商户和小程序商户。对这些商户主体的风险管理包含了商户准入、商户风险识别、商户风险运营和商户风险决策者四个主要环节。
在商户管理的各个环节中，风控运营、策略和模型都是基于商户主体的各种数据，而实际上，不同商户主体之间存在千丝万缕的联系(例如身份重合关系、资金往来关系、介质共用关系、地理位置相近等)，如果能够基于主体之间的联系，将不同的主体聚合在一起，打通不同主体的数据，对每个主体进行分析、管控和建模，则能够有效地提升风险管理的准确性、覆盖率和实效性，实现生态风险联防。
聚合计算中一个重要的环节是分位数的计算。分位数的计算在描述性统计中很常见，比如说相较于平均数、中位数不会受到异常值的影响，但分位数的计算过程比较复杂，需要保留所有具体值，排序后取得中间位置的数作为结果。这种复杂的分位数计算方法导致主体聚合效率降低，以及导致主体管理(如主体风险防控等)效率降低等。
发明内容
一方面，本说明书一个或多个实施例提供一种资源数据的处理方法，包括：基于多个资源数据的分割位置，确定所述多个资源数据对应的至少一个资源分割值。所述资源分割值与所述资源数据的极端值之间的第一差异度小于所述分割位置与所述极端值之间的第二差异度。所述极端值包括所述资源数据中的最大数据值和/或最小数据值。利用各所述资源分割值对所述多个资源数据进行聚类处理，得到多个资源聚类组。根据预设的资源评估指标，从所述多个资源聚类组中确定所述资源评估指标对应的目标资源聚类组。所述资源评估指标包含对所述资源数据进行风险评估所使用的风险评估参数。根据所述目标资源聚类组对应的目标资源分割值，确定所述资源数据对应的资源评估阈值。所述资源评估阈值用于对所述资源数据进行风险评估。
另一方面，本说明书一个或多个实施例提供一种资源数据的处理装置，包括：第一确定模块，基于多个资源数据的分割位置，确定所述多个资源数据对应的至少一个资 源分割值。所述资源分割值与所述资源数据的极端值之间的第一差异度小于所述分割位置与所述极端值之间的第二差异度。所述极端值包括所述资源数据中的最大数据值和/或最小数据值。聚类模块，利用各所述资源分割值对所述多个资源数据进行聚类处理，得到多个资源聚类组。第二确定模块，根据预设的资源评估指标，从所述多个资源聚类组中确定所述资源评估指标对应的目标资源聚类组。所述资源评估指标包含对所述资源数据进行风险评估所使用的风险评估参数。第三确定模块，根据所述目标资源聚类组对应的目标资源分割值，确定所述资源数据对应的资源评估阈值。所述资源评估阈值用于对所述资源数据进行风险评估。
再一方面，本说明书一个或多个实施例提供一种资源数据的处理设备，包括：处理器；以及被安排成存储计算机可执行指令的存储器，所述可执行指令在被执行时使所述处理器：基于多个资源数据的分割位置，确定所述多个资源数据对应的至少一个资源分割值。所述资源分割值与所述资源数据的极端值之间的第一差异度小于所述分割位置与所述极端值之间的第二差异度。所述极端值包括所述资源数据中的最大数据值和/或最小数据值。利用各所述资源分割值对所述多个资源数据进行聚类处理，得到多个资源聚类组。根据预设的资源评估指标，从所述多个资源聚类组中确定所述资源评估指标对应的目标资源聚类组。所述资源评估指标包含对所述资源数据进行风险评估所使用的风险评估参数。根据所述目标资源聚类组对应的目标资源分割值，确定所述资源数据对应的资源评估阈值。所述资源评估阈值用于对所述资源数据进行风险评估。
再一方面，本说明书一个或多个实施例提供一种存储介质，用于存储计算机可执行指令，所述可执行指令在被执行时实现以下流程：基于多个资源数据的分割位置，确定所述多个资源数据对应的至少一个资源分割值。所述资源分割值与所述资源数据的极端值之间的第一差异度小于所述分割位置与所述极端值之间的第二差异度。所述极端值包括所述资源数据中的最大数据值和/或最小数据值。利用各所述资源分割值对所述多个资源数据进行聚类处理，得到多个资源聚类组。根据预设的资源评估指标，从所述多个资源聚类组中确定所述资源评估指标对应的目标资源聚类组。所述资源评估指标包含对所述资源数据进行风险评估所使用的风险评估参数。根据所述目标资源聚类组对应的目标资源分割值，确定所述资源数据对应的资源评估阈值。所述资源评估阈值用于对所述资源数据进行风险评估。
附图说明
为了更清楚地说明本说明书一个或多个实施例或现有技术中的技术方案，下面将 对实施例或现有技术描述中所需要使用的附图作简单地介绍，显见地，下面描述中的附图仅仅是本说明书一个或多个实施例中记载的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1是根据本说明书一实施例的一种资源数据的处理方法的示意性流程图；
图2是根据本说明书另一实施例的一种资源数据的处理方法的示意性流程图；
图3是根据本说明书一实施例的一种资源数据的处理装置的示意性框图；
图4是根据本说明书一实施例的一种资源数据的处理设备的示意性框图。
具体实施方式
本说明书一个或多个实施例提供一种资源数据的处理方法及装置，用以解决数据聚类效率低以及风险管理效率低的问题。
为了使本技术领域的人员更好地理解本说明书一个或多个实施例中的技术方案，下面将结合本说明书一个或多个实施例中的附图，对本说明书一个或多个实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本说明书一部分实施例，而不是全部的实施例。基于本说明书一个或多个实施例，本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都应当属于本文件保护的范围。
图1是根据本说明书一实施例的一种资源数据的处理方法的示意性流程图，如图1所示，该方法包括步骤S102～S108。
S102，基于多个资源数据的分割位置，确定多个资源数据对应的至少一个资源分割值。
其中，资源分割值与资源数据的极端值之间的第一差异度小于分割位置与极端值之间的第二差异度。极端值包括资源数据中的最大数据值和/或最小数据值。例如，将资源数据按照数据大小的顺序进行排列，那么极端值即为排在最前的资源数据和排在最后的资源数据。资源分割值一般位于0～1之间，其表示对资源数据的分割位置。例如，资源分割值为20％，则表示在资源数据的20％位置处进行分割。
资源分割值与资源数据的极端值之间的第一差异度，即为资源分割值与资源数据的的极端值的接近程度。分割数据与资源数据的极端值之间的第二差异度，即为分割数据与资源数据的的极端值的接近程度。
S104，利用各资源分割值对多个资源数据进行聚类处理，得到多个资源聚类组。
S106，根据预设的资源评估指标，从多个资源聚类组中确定资源评估指标对应的目标资源聚类组。
其中，资源评估指标包含对资源数据进行风险评估所使用的风险评估参数。风险评估参数例如：低风险数据在所有资源数据中的占比。例如，风险评估参数为低风险数据在所有资源数据中的占比90％，则说明预设的资源评估指标希望对资源数据进行风险评估时低风险数据的占比为90％。当然，还可预设其他资源评估指标，例如低风险数据的总数目或高风险数据的总数目等。
S108，根据目标资源聚类组对应的目标资源分割值，确定资源数据对应的资源评估阈值，资源评估阈值用于对资源数据进行风险评估。
采用本说明书一个或多个实施例的技术方案，能够基于多个资源数据的分割位置确定多个资源数据对应的至少一个资源分割值，并利用各资源分割值对多个资源数据进行聚类处理，得到多个资源聚类组。可见，对资源数据进行分割时并不是基于原始的分割位置分割的，而是通过计算后的与资源数据的极端值之间的差异度更小的资源分割值进行分割，由于实际应用中人们更关心处于极端位置的资源数据，因此可使得资源数据的聚类结果更加精准、快速。并且通过根据预设的资源评估指标从多个资源聚类组中确定资源评估指标对应的目标资源聚类组，进而根据目标资源聚类组对应的目标资源分割值确定资源数据对应的资源评估阈值，使得资源评估阈值能够基于预设的资源评估指标来确定，不仅提升了资源评估阈值的精确度，且使得资源数据的风险评估结果更加准确。
上述实施例提供的资源数据的处理方法中，对资源数据的聚类可利用Tdigest算法。TDigest是一个简单、快速、精确度高、可并行化的近似百分位算法，TDigest算法的核心思想是Sketch(素描)，是一种抽象、简化的思想，一种处理问题的方式。以下详细介绍如何利用Tdigest算法对资源数据进行聚类。
在一个实施例中，执行S102时，可先确定对多个资源数据进行分割的至少一个分割位置；进而根据各分割位置与资源分割值之间的映射关系，计算各分割位置分别对应的资源分割值。
其中，分割位置可以是百分数的形式，分割位置对应的资源分割值可以为资源数据的质心。分割位置与资源分割值之间的映射关系(即百分数与质心之间的映射关系)如下述公式(1)。
Figure PCTCN2020132620-appb-000001
其中，q表示百分数；k表示质心；δ为常量，通常取比较小的值，如0.01。δ的取值影响质心k的大小。
本实施例中，通过各分割位置与资源分割值之间的映射关系即可将分割位置转化为资源分割值，由于资源分割值与资源数据的极端值之间的差异度更小，因此能够使资源数据的聚类结果更加精准、快速。
在一个实施例中，执行S104时，可首先将各资源分割值按照预设维度进行排序，预设维度包括资源数据的数据大小。然后基于排序后的各资源分割值，将每两个相邻的资源分割值确定为一个资源聚类组对应的边界值。再基于各资源聚类组分别对应的边界值，确定多个资源聚类组，将各资源数据分别划分至对应的各资源聚类组中。
假设资源分割值为资源数据的质心，对各资源分割值按照资源数据的大小进行排序，即按照质心大小对各质心进行排序。然后将每两个相邻的质心确定为一个资源聚类组对应的边界值，再基于每个资源聚类组分别对应的边界值确定出多个资源聚类组，进而将各资源数据分别划分至对应的各资源聚类组中。
将各资源数据分别划分至对应的各资源聚类组中时，可将各资源数据按照数据大小进行排序，然后将各资源数据与各资源聚类组分别对应的边界值进行比对，以确定出各资源数据落入哪一个资源聚类组，并将各资源数据划分至各自所落入的资源聚类组内。
本实施例中，由于依据资源分割值(即质心)对资源数据进行聚类，且资源分割值更加靠近资源数据的极端值，因此聚类结果更加精确、更加符合实际应用中人们对极端位置资源数据的关心需求。
上述实施例中指出，风险评估参数可以为高风险数据在所有资源数据中的占比。基于此，若风险评估参数包含第一资源分割值，即以第一资源分割值表征高风险数据在所有资源数据中的占比。则执行S106时，可从多个资源聚类组中确定出第一资源分割值所在的资源聚类组，并将第一资源分割值所在的资源聚类组确定为目标资源聚类组。
在一个实施例中，根据目标资源聚类组对应的目标资源分割值确定资源数据对应的资源评估阈值时，可首先确定各目标资源分割值分别对应的权重，进而根据各目标资源分割值及各目标资源分割值分别对应的权重，计算资源数据对应的资源评估阈值。其中，各目标资源分割值分别对应的权重之和为1。目标资源聚类组对应的目标资源分割值即为目标资源聚类组的两个边界值。
本实施例中，若风险评估参数包含第一资源分割值，即以第一资源分割值表征高 风险数据在所有资源数据中的占比。目标资源聚类组包含第一目标资源分割值及第二目标资源分割值(即目标资源聚类组的两个边界值)。则可按照如下方式确定各目标资源分割值分别对应的权重：首先，确定第一资源分割值对应的资源百分位数。其次，确定第一资源分割值对应的资源百分位数为第一目标资源分割值对应的权重。再次，计算资源百分位数与1之间的差值的绝对值，确定该绝对值为第二目标资源分割值对应的权重。
上述实施例中，通过根据预设的资源评估指标从多个资源聚类组中确定资源评估指标对应的目标资源聚类组，进而根据目标资源聚类组对应的目标资源分割值确定资源数据对应的资源评估阈值，使得资源评估阈值能够基于预设的资源评估指标来确定，不仅提升了资源评估阈值的精确度，且使得资源数据的风险评估结果更加准确。
上述实施例提供的资源数据的处理方法可应用于各类资源数据的风险评估场景中。例如，对交易金额的风险评估、业务数据的风险评估等。下面以对交易金额的风险评估场景为例进行说明。
图2是根据本说明书另一实施例的一种资源数据的处理方法的示意性流程图。该实施例中，资源数据为某一时间段内的多笔交易金额。如图2所示，该方法包括步骤S201～S208。
S201，确定对多笔交易金额进行分割的多个百分数。
其中，百分数即为多笔交易金额的分割位置，例如，确定对多笔交易金额进行分割的百分数为20％、40％、60％和80％。百分数的确定可由用户指定，也可由计算机按照预设规则来确定。预设规则可以是以下任一种规则：从0～1之间随机确定百分数、从0～1之间按照预设间隔均匀确定百分数、从0～1之间均匀确定N个百分数，等等。
S202，基于确定的多个百分数，确定多笔交易金额对应的多个金额分割值。
其中，金额分割值相当于多笔交易金额的质心。已知百分数，则可通过上述公式(1)来确定出每个百分数对应的金额分割值。该步骤确定出的金额分割值一般位于0～1之间，其表示对交易金额的分割位置。例如，金额分割值为20％，则表示在交易金额的20％位置处进行分割。
本实施例中，质心具有以下特征：质心与交易金额的极端值之间的第一差异度小于S201中确定的百分数与交易金额的极端值之间的第二差异度。其中，交易金额的极端值包括所有交易金额中的最大金额值和最小金额值。即，将百分数转化为质心对交易金额进行分割，可使得交易金额的分割位置更靠近交易金额的极端值。
S203，对多个金额分割值按照金额大小进行排序，并将每两个相邻的金额分割值作为一个金额聚类组对应的边界值，得到多个金额聚类组。
S204，将各笔交易金额分别划分至对应的金额聚类组中。
该步骤中，可将各笔交易金额按照金额大小进行排序，然后看每笔交易金额落入哪一个金额聚类组，再将每笔交易金额划分至各自所落入的金额聚类组中。
S205，确定交易金额评估指标。
其中，交易金额评估指标包括对交易金额进行风险评估所使用的风险评估参数，例如，低风险交易金额在所有交易金额中的占比。
S206，从多个金额聚类组中确定交易金额评估指标对应的目标金额聚类组。
假设交易金额评估指标为低风险交易金额在所有交易金额中的占比90％。再假设S202确定出的多个金额分割值分别为：0.2、0.8、0.95。则可确定交易金额评估指标对应的目标金额聚类组为由金额分割值0.8和0.95组成的金额聚类组，即将金额分割值0.8和0.95作为两个边界值的金额聚类组。
S207，根据交易金额评估指标以及目标金额聚类组对应的目标金额分割值，并利用插值法计算出金额评估阈值。
该步骤中，利用插值法计算出金额评估阈值的具体方式如下：首先，确定金额百分位数(即交易金额评估指标)。其次，确定金额百分位数为第一目标金额分割值(即目标金额聚类组对应的较小的目标金额分割值)对应的权重，以及，计算金额百分位数与1之间的差值的绝对值，确定该绝对值为第二目标金额分割值(即目标金额聚类组对应的较大的目标金额分割值)对应的权重。最后，基于各目标金额分割值及各目标金额分割值分别对应的权重，计算交易金额对应的金额评估阈值。
假设金额百分位数为90％，则第一目标金额分割值为a，其对应的权重为0.9；第二目标金额分割值为b，其对应的权重为0.1。金额评估阈值为：a*0.9+b*0.1。
S208，基于金额评估阈值对交易金额进行风险评估。
例如，若交易金额大于金额评估阈值，则可确定交易金额属于高风险交易金额；若交易金额小于或等于金额评估阈值，则可确定交易金额属于低风险交易金额。
本实施例中，能够基于与交易金额的极端值之间的差异度更小的各金额分割值对多笔交易金额进行聚类处理，得到多个金额聚类组。由于实际应用中人们更关心处于极 端位置的交易金额的风险情况，因此可使得交易金额的聚类结果更加精准、快速。并且通过根据预设的交易金额评估指标从多个金额聚类组中确定交易金额评估指标对应的目标金额聚类组，进而根据目标金额聚类组对应的目标金额分割值确定交易金额对应的金额评估阈值，使得金额评估阈值能够基于预设的交易金额评估指标来确定，不仅提升了交易金额评估阈值的精确度，且使得交易金额的风险评估结果更加准确。
上述对本说明书特定实施例进行了描述。其它实施例在所附权利要求书的范围内。在一些情况下，在权利要求书中记载的动作或步骤可以按照不同于实施例中的顺序来执行并且仍然可以实现期望的结果。另外，在附图中描绘的过程不一定要求示出的特定顺序或者连续顺序才能实现期望的结果。在某些实施方式中，多任务处理和并行处理也是可以的或者可能是有利的。
以上为本说明书一个或多个实施例提供的资源数据的处理方法，基于同样的思路，本说明书一个或多个实施例还提供一种资源数据的处理装置。
图3是根据本说明书一实施例的一种资源数据的处理装置的示意性框图。如图3所示，资源数据的处理装置包括：第一确定模块310、聚类模块320、确定模块330、三确定模块340。
第一确定模块310，基于多个资源数据的分割位置，确定多个资源数据对应的至少一个资源分割值；资源分割值与资源数据的极端值之间的第一差异度小于分割位置与极端值之间的第二差异度；极端值包括资源数据中的最大数据值和/或最小数据值；
聚类模块320，利用各资源分割值对多个资源数据进行聚类处理，得到多个资源聚类组；
第二确定模块330，根据预设的资源评估指标，从多个资源聚类组中确定资源评估指标对应的目标资源聚类组；资源评估指标包含对资源数据进行风险评估所使用的风险评估参数；
第三确定模块340，根据目标资源聚类组对应的目标资源分割值，确定资源数据对应的资源评估阈值；资源评估阈值用于对资源数据进行风险评估。
在一个实施例中，第一确定模块310包括：第一确定单元，确定对多个资源数据进行分割的至少一个分割位置；计算单元，根据各分割位置与资源分割值之间的映射关系，计算各分割位置分别对应的资源分割值。
在一个实施例中，聚类模块320包括：排序单元，将各资源分割值按照预设维度 进行排序；预设维度包括资源数据的数据大小；第二确定单元，基于排序后的各资源分割值，将每两个相邻的资源分割值确定为一个资源聚类组对应的边界值；第三确定单元，基于各资源聚类组分别对应的边界值，确定多个资源聚类组；
划分单元，将各资源数据分别划分至对应的各资源聚类组中。
在一个实施例中，第三确定模块340包括：第四确定单元，确定各目标资源分割值分别对应的权重；其中，各目标资源分割值分别对应的权重之和为1；计算单元，根据各目标资源分割值及各目标资源分割值分别对应的权重，计算资源数据对应的资源评估阈值。
在一个实施例中，风险评估参数包含第一资源分割值；目标资源聚类组包含第一目标资源分割值及第二目标资源分割值。
第四确定单元还用于：确定第一资源分割值对应的资源百分位数；确定第一资源分割值对应的资源百分位数为第一目标资源分割值对应的权重；计算资源百分位数与1之间的差值的绝对值；确定绝对值为第二目标资源分割值对应的权重。
在一个实施例中，第二确定模块330包括：第五确定单元，从多个资源聚类组中确定第一资源分割值所在的资源聚类组；第六确定单元，将第一资源分割值所在的资源聚类组确定为目标资源聚类组。
采用本说明书一个或多个实施例的装置，能够基于多个资源数据的分割位置确定多个资源数据对应的至少一个资源分割值，并利用各资源分割值对多个资源数据进行聚类处理，得到多个资源聚类组。可见，对资源数据进行分割时并不是基于原始的分割位置分割的，而是通过计算后的与资源数据的极端值之间的差异度更小的资源分割值进行分割，由于实际应用中人们更关心处于极端位置的资源数据，因此可使得资源数据的聚类结果更加精准、快速。并且通过根据预设的资源评估指标从多个资源聚类组中确定资源评估指标对应的目标资源聚类组，进而根据目标资源聚类组对应的目标资源分割值确定资源数据对应的资源评估阈值，使得资源评估阈值能够基于预设的资源评估指标来确定，不仅提升了资源评估阈值的精确度，且使得资源数据的风险评估结果更加准确。
本领域的技术人员应可理解，上述资源数据的处理装置能够用来实现前文所述的资源数据的处理方法，其中的细节描述应与前文方法部分描述类似，为避免繁琐，此处不另赘述。
基于同样的思路，本说明书一个或多个实施例还提供一种资源数据的处理设备， 如图4所示。资源数据的处理设备可因配置或性能不同而产生比较大的差异，可以包括一个或一个以上的处理器401和存储器402，存储器402中可以存储有一个或一个以上存储应用程序或数据。其中，存储器402可以是短暂存储或持久存储。存储在存储器402的应用程序可以包括一个或一个以上模块(图示未示出)，每个模块可以包括对资源数据的处理设备中的一系列计算机可执行指令。更进一步地，处理器401可以设置为与存储器402通信，在资源数据的处理设备上执行存储器402中的一系列计算机可执行指令。资源数据的处理设备还可以包括一个或一个以上电源403，一个或一个以上有线或无线网络接口404，一个或一个以上输入输出接口405，一个或一个以上键盘406。
具体在本实施例中，资源数据的处理设备包括有存储器，以及一个或一个以上的程序，其中一个或者一个以上程序存储于存储器中，且一个或者一个以上程序可以包括一个或一个以上模块，且每个模块可以包括对资源数据的处理设备中的一系列计算机可执行指令，且经配置以由一个或者一个以上处理器执行该一个或者一个以上程序包含用于进行以下计算机可执行指令：基于多个资源数据的分割位置，确定所述多个资源数据对应的至少一个资源分割值；所述资源分割值与所述资源数据的极端值之间的第一差异度小于所述分割位置与所述极端值之间的第二差异度；所述极端值包括所述资源数据中的最大数据值和/或最小数据值；利用各所述资源分割值对所述多个资源数据进行聚类处理，得到多个资源聚类组；根据预设的资源评估指标，从所述多个资源聚类组中确定所述资源评估指标对应的目标资源聚类组；所述资源评估指标包含对所述资源数据进行风险评估所使用的风险评估参数；根据所述目标资源聚类组对应的目标资源分割值，确定所述资源数据对应的资源评估阈值；所述资源评估阈值用于对所述资源数据进行风险评估。
可选地，计算机可执行指令在被执行时，还可以使所述处理器：确定对所述多个资源数据进行分割的至少一个所述分割位置；根据各所述分割位置与所述资源分割值之间的映射关系，计算各所述分割位置分别对应的所述资源分割值。
可选地，计算机可执行指令在被执行时，还可以使所述处理器：将各所述资源分割值按照预设维度进行排序；所述预设维度包括所述资源数据的数据大小；基于排序后的各所述资源分割值，将每两个相邻的所述资源分割值确定为一个资源聚类组对应的边界值；基于各所述资源聚类组分别对应的所述边界值，确定多个所述资源聚类组；将各所述资源数据分别划分至对应的各所述资源聚类组中。
可选地，计算机可执行指令在被执行时，还可以使所述处理器：确定各所述目标 资源分割值分别对应的权重；其中，各所述目标资源分割值分别对应的权重之和为1；根据各所述目标资源分割值及各所述目标资源分割值分别对应的权重，计算所述资源数据对应的所述资源评估阈值。
可选地，所述风险评估参数包含第一资源分割值；所述目标资源聚类组包含第一目标资源分割值及第二目标资源分割值；计算机可执行指令在被执行时，还可以使所述处理器：确定所述第一资源分割值对应的资源百分位数；确定所述第一资源分割值对应的资源百分位数为所述第一目标资源分割值对应的权重；计算所述资源百分位数与1之间的差值的绝对值；确定所述绝对值为所述第二目标资源分割值对应的权重。
可选地，计算机可执行指令在被执行时，还可以使所述处理器：从所述多个资源聚类组中确定所述第一资源分割值所在的资源聚类组；将所述第一资源分割值所在的资源聚类组确定为所述目标资源聚类组。
本说明书一个或多个实施例还提出了一种计算机可读存储介质，该计算机可读存储介质存储一个或多个程序，该一个或多个程序包括指令，该指令当被包括多个应用程序的电子设备执行时，能够使该电子设备执行上述资源数据的处理方法，并具体用于执行：基于多个资源数据的分割位置，确定所述多个资源数据对应的至少一个资源分割值；所述资源分割值与所述资源数据的极端值之间的第一差异度小于所述分割位置与所述极端值之间的第二差异度；所述极端值包括所述资源数据中的最大数据值和/或最小数据值；利用各所述资源分割值对所述多个资源数据进行聚类处理，得到多个资源聚类组；根据预设的资源评估指标，从所述多个资源聚类组中确定所述资源评估指标对应的目标资源聚类组；所述资源评估指标包含对所述资源数据进行风险评估所使用的风险评估参数；根据所述目标资源聚类组对应的目标资源分割值，确定所述资源数据对应的资源评估阈值；所述资源评估阈值用于对所述资源数据进行风险评估。
上述实施例阐明的系统、装置、模块或单元，具体可以由计算机芯片或实体实现，或者由具有某种功能的产品来实现。一种典型的实现设备为计算机。具体的，计算机例如可以为个人计算机、膝上型计算机、蜂窝电话、相机电话、智能电话、个人数字助理、媒体播放器、导航设备、电子邮件设备、游戏控制台、平板计算机、可穿戴设备或者这些设备中的任何设备的组合。
为了描述的方便，描述以上装置时以功能分为各种单元分别描述。当然，在实施本说明书一个或多个实施例时可以把各单元的功能在同一个或多个软件和/或硬件中实现。
本领域内的技术人员应明白，本说明书一个或多个实施例可提供为方法、系统、或计算机程序产品。因此，本说明书一个或多个实施例可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且，本说明书一个或多个实施例可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。
本说明书一个或多个实施例是参照根据本申请实施例的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器，使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理设备以特定方式工作的计算机可读存储器中，使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品，该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理设备上，使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理，从而在计算机或其他可编程设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
在一个典型的配置中，计算设备包括一个或多个处理器(CPU)、输入/输出接口、网络接口和内存。
内存可能包括计算机可读介质中的非永久性存储器，随机存取存储器(RAM)和/或非易失性内存等形式，如只读存储器(ROM)或闪存(flash RAM)。内存是计算机可读介质的示例。
计算机可读介质包括永久性和非永久性、可移动和非可移动媒体可以由任何方法或技术来实现信息存储。信息可以是计算机可读指令、数据结构、程序的模块或其他数据。计算机的存储介质的例子包括，但不限于相变内存(PRAM)、静态随机存取存储器(SRAM)、动态随机存取存储器(DRAM)、其他类型的随机存取存储器(RAM)、 只读存储器(ROM)、电可擦除可编程只读存储器(EEPROM)、快闪记忆体或其他内存技术、只读光盘只读存储器(CD-ROM)、数字多功能光盘(DVD)或其他光学存储、磁盒式磁带，磁带磁磁盘存储或其他磁性存储设备或任何其他非传输介质，可用于存储可以被计算设备访问的信息。按照本文中的界定，计算机可读介质不包括暂存电脑可读媒体(transitory media)，如调制的数据信号和载波。
还需要说明的是，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、商品或者设备不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、商品或者设备所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括所述要素的过程、方法、商品或者设备中还存在另外的相同要素。
本说明书一个或多个实施例可以在由计算机执行的计算机可执行指令的一般上下文中描述，例如程序模块。一般地，程序模块包括执行特定任务或实现特定抽象数据类型的例程、程序、对象、组件、数据结构等等。也可以在分布式计算环境中实践本申请，在这些分布式计算环境中，由通过通信网络而被连接的远程处理设备来执行任务。在分布式计算环境中，程序模块可以位于包括存储设备在内的本地和远程计算机存储介质中。
本说明书中的各个实施例均采用递进的方式描述，各个实施例之间相同相似的部分互相参见即可，每个实施例重点说明的都是与其他实施例的不同之处。尤其，对于系统实施例而言，由于其基本相似于方法实施例，所以描述的比较简单，相关之处参见方法实施例的部分说明即可。
以上所述仅为本说明书一个或多个实施例而已，并不用于限制本说明书。对于本领域技术人员来说，本说明书一个或多个实施例可以有各种更改和变化。凡在本说明书一个或多个实施例的精神和原理之内所作的任何修改、等同替换、改进等，均应包含在本说明书一个或多个实施例的权利要求范围之内。

Claims (11)

  1. 一种资源数据的处理方法，包括：
    基于多个资源数据的分割位置，确定所述多个资源数据对应的至少一个资源分割值；所述资源分割值与所述资源数据的极端值之间的第一差异度小于所述分割位置与所述极端值之间的第二差异度；所述极端值包括所述资源数据中的最大数据值和/或最小数据值；
    利用各所述资源分割值对所述多个资源数据进行聚类处理，得到多个资源聚类组；
    根据预设的资源评估指标，从所述多个资源聚类组中确定所述资源评估指标对应的目标资源聚类组；所述资源评估指标包含对所述资源数据进行风险评估所使用的风险评估参数；
    根据所述目标资源聚类组对应的目标资源分割值，确定所述资源数据对应的资源评估阈值；所述资源评估阈值用于对所述资源数据进行风险评估。
  2. 根据权利要求1所述的方法，基于多个资源数据的分割位置，确定所述多个资源数据对应的至少一个资源分割值，包括：
    确定对所述多个资源数据进行分割的至少一个所述分割位置；
    根据各所述分割位置与所述资源分割值之间的映射关系，计算各所述分割位置分别对应的所述资源分割值。
  3. 根据权利要求1所述的方法，利用各所述资源分割值对所述多个资源数据进行聚类处理，得到多个资源聚类组，包括：
    将各所述资源分割值按照预设维度进行排序；所述预设维度包括所述资源数据的数据大小；
    基于排序后的各所述资源分割值，将每两个相邻的所述资源分割值确定为一个资源聚类组对应的边界值；
    基于各所述资源聚类组分别对应的所述边界值，确定多个所述资源聚类组；
    将各所述资源数据分别划分至对应的各所述资源聚类组中。
  4. 根据权利要求1所述的方法，根据所述目标资源聚类组对应的目标资源分割值，确定所述资源数据对应的资源评估阈值，包括：
    确定各所述目标资源分割值分别对应的权重；其中，各所述目标资源分割值分别对应的权重之和为1；
    根据各所述目标资源分割值及各所述目标资源分割值分别对应的权重，计算所述资源数据对应的所述资源评估阈值。
  5. 根据权利要求4所述的方法，所述风险评估参数包含第一资源分割值；所述目标资源聚类组包含第一目标资源分割值及第二目标资源分割值；
    所述确定各所述目标资源分割值分别对应的权重，包括：
    确定所述第一资源分割值对应的资源百分位数；
    确定所述第一资源分割值对应的资源百分位数为所述第一目标资源分割值对应的权重；
    计算所述资源百分位数与1之间的差值的绝对值；确定所述绝对值为所述第二目标资源分割值对应的权重。
  6. 根据权利要求5所述的方法，根据预设的资源评估指标，从所述多个资源聚类组中确定所述资源评估指标对应的目标资源聚类组，包括：
    从所述多个资源聚类组中确定所述第一资源分割值所在的资源聚类组；
    将所述第一资源分割值所在的资源聚类组确定为所述目标资源聚类组。
  7. 一种资源数据的处理装置，包括：
    第一确定模块，基于多个资源数据的分割位置，确定所述多个资源数据对应的至少一个资源分割值；所述资源分割值与所述资源数据的极端值之间的第一差异度小于所述分割位置与所述极端值之间的第二差异度；所述极端值包括所述资源数据中的最大数据值和/或最小数据值；
    聚类模块，利用各所述资源分割值对所述多个资源数据进行聚类处理，得到多个资源聚类组；
    第二确定模块，根据预设的资源评估指标，从所述多个资源聚类组中确定所述资源评估指标对应的目标资源聚类组；所述资源评估指标包含对所述资源数据进行风险评估所使用的风险评估参数；
    第三确定模块，根据所述目标资源聚类组对应的目标资源分割值，确定所述资源数据对应的资源评估阈值；所述资源评估阈值用于对所述资源数据进行风险评估。
  8. 根据权利要求7所述的装置，所述第一确定模块包括：
    第一确定单元，确定对所述多个资源数据进行分割的至少一个所述分割位置；
    计算单元，根据各所述分割位置与所述资源分割值之间的映射关系，计算各所述分割位置分别对应的所述资源分割值。
  9. 根据权利要求7所述的装置，所述聚类模块包括：
    排序单元，将各所述资源分割值按照预设维度进行排序；所述预设维度包括所述资源数据的数据大小；
    第二确定单元，基于排序后的各所述资源分割值，将每两个相邻的所述资源分割值确定为一个资源聚类组对应的边界值；
    第三确定单元，基于各所述资源聚类组分别对应的所述边界值，确定多个所述资源聚类组；
    划分单元，将各所述资源数据分别划分至对应的各所述资源聚类组中。
  10. 一种资源数据的处理设备，包括：
    处理器；以及
    被安排成存储计算机可执行指令的存储器，所述可执行指令在被执行时使所述处理器：
    基于多个资源数据的分割位置，确定所述多个资源数据对应的至少一个资源分割值；所述资源分割值与所述资源数据的极端值之间的第一差异度小于所述分割位置与所述极端值之间的第二差异度；所述极端值包括所述资源数据中的最大数据值和/或最小数据值；
    利用各所述资源分割值对所述多个资源数据进行聚类处理，得到多个资源聚类组；
    根据预设的资源评估指标，从所述多个资源聚类组中确定所述资源评估指标对应的目标资源聚类组；所述资源评估指标包含对所述资源数据进行风险评估所使用的风险评估参数；
    根据所述目标资源聚类组对应的目标资源分割值，确定所述资源数据对应的资源评估阈值；所述资源评估阈值用于对所述资源数据进行风险评估。
  11. 一种存储介质，用于存储计算机可执行指令，所述可执行指令在被执行时实现以下流程：
    基于多个资源数据的分割位置，确定所述多个资源数据对应的至少一个资源分割值；所述资源分割值与所述资源数据的极端值之间的第一差异度小于所述分割位置与所述极端值之间的第二差异度；所述极端值包括所述资源数据中的最大数据值和/或最小数据值；
    利用各所述资源分割值对所述多个资源数据进行聚类处理，得到多个资源聚类组；
    根据预设的资源评估指标，从所述多个资源聚类组中确定所述资源评估指标对应的目标资源聚类组；所述资源评估指标包含对所述资源数据进行风险评估所使用的风险评估参数；
    根据所述目标资源聚类组对应的目标资源分割值，确定所述资源数据对应的资源评估阈值；所述资源评估阈值用于对所述资源数据进行风险评估。
PCT/CN2020/132620 2020-01-14 2020-11-30 资源数据的处理方法及装置 Ceased WO2021143370A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010035782.8A CN111310784B (zh) 2020-01-14 2020-01-14 资源数据的处理方法及装置
CN202010035782.8 2020-01-14

Publications (1)

Publication Number Publication Date
WO2021143370A1 true WO2021143370A1 (zh) 2021-07-22

Family

ID=71156406

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/132620 Ceased WO2021143370A1 (zh) 2020-01-14 2020-11-30 资源数据的处理方法及装置

Country Status (2)

Country Link
CN (2) CN113780404B (zh)
WO (1) WO2021143370A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113706336A (zh) * 2021-09-01 2021-11-26 知能汇融(北京)咨询有限公司 风险评估方法、装置、计算机设备和存储介质

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113780404B (zh) * 2020-01-14 2024-11-26 支付宝(杭州)信息技术有限公司 资源数据的处理方法及装置
CN113344104A (zh) * 2021-06-23 2021-09-03 支付宝(杭州)信息技术有限公司 一种数据处理方法、装置、设备及介质
CN114070895B (zh) * 2021-11-15 2023-04-25 中国联合网络通信集团有限公司 数据传输方法、控制面网元和用户面网元
CN115442262B (zh) * 2022-08-01 2024-02-06 阿里巴巴（中国）有限公司 一种资源评估方法、装置、电子设备及存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103297267A (zh) * 2013-05-10 2013-09-11 河北远东通信系统工程有限公司 一种网络行为的风险评估方法和系统
CN106779723A (zh) * 2016-12-26 2017-05-31 中国银联股份有限公司 一种移动终端风险评估方法及装置
CN109598414A (zh) * 2018-11-13 2019-04-09 阿里巴巴集团控股有限公司 风险评估模型训练、风险评估方法、装置及电子设备
WO2019222017A1 (en) * 2018-05-18 2019-11-21 Beckman Coulter, Inc. Test panel analysis
CN111310784A (zh) * 2020-01-14 2020-06-19 支付宝(杭州)信息技术有限公司 资源数据的处理方法及装置

Family Cites Families (29)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103107903B (zh) * 2011-11-15 2016-01-27 中国移动通信集团广东有限公司 一种资源数据共享方法和资源数据共享设备
US20140278733A1 (en) * 2013-03-15 2014-09-18 Navin Sabharwal Risk management methods and systems for enterprise processes
CN103426056B (zh) * 2013-07-18 2016-06-29 清华大学 基于风险评估的电力系统薄弱环节辨识方法
CN103530347B (zh) * 2013-10-09 2016-09-14 北京东方网信科技股份有限公司 一种基于大数据挖掘的互联网资源质量评估方法及系统
CN104869166A (zh) * 2015-05-28 2015-08-26 北京呈创科技股份有限公司 桌面云集群系统及桌面云集群中的信息交互方法
CN105045806B (zh) * 2015-06-04 2019-04-09 中国科学院信息工程研究所 一种面向分位数查询的概要数据动态分裂与维护方法
US20180300829A1 (en) * 2015-10-28 2018-10-18 Fractal Industries, Inc. System for intellectual property landscape analysis, risk management, and opportunity identification
CN107563757B (zh) * 2016-07-01 2020-09-22 阿里巴巴集团控股有限公司 数据风险识别的方法及装置
CN106202477A (zh) * 2016-07-18 2016-12-07 北京千安哲信息技术有限公司 医疗费用挖掘方法及装置
CN107645533A (zh) * 2016-07-22 2018-01-30 阿里巴巴集团控股有限公司 数据处理方法、数据发送方法、风险识别方法及设备
CN108399477B (zh) * 2017-02-07 2021-03-05 创新先进技术有限公司 一种风险阈值确定的方法及装置
CN107194016A (zh) * 2017-07-14 2017-09-22 上海瀚之友信息技术服务有限公司 一种数据分割方法
CN107909233A (zh) * 2017-08-31 2018-04-13 上海壹账通金融科技有限公司 风控数据统计方法及应用服务器
CN107704872A (zh) * 2017-09-19 2018-02-16 安徽理工大学 一种基于相对最离散维分割的K‑means聚类初始中心选取方法
CN107679946B (zh) * 2017-09-28 2021-09-10 平安科技（深圳）有限公司 基金产品推荐方法、装置、终端设备及存储介质
CN107977768A (zh) * 2017-10-25 2018-05-01 阿里巴巴集团控股有限公司 一种可增值资源的评估方法及装置
CN108549954B (zh) * 2018-03-26 2022-08-02 平安科技（深圳）有限公司 风险模型训练方法、风险识别方法、装置、设备及介质
CN110458376A (zh) * 2018-05-07 2019-11-15 上海诺悦智能科技有限公司 一种可疑风险交易筛查方法与相应系统
CN108711004A (zh) * 2018-05-14 2018-10-26 北京京东金融科技控股有限公司 风险评估方法和装置及计算机可读存储介质
CN108959934B (zh) * 2018-06-11 2023-08-22 平安科技（深圳）有限公司 安全风险评估方法、装置、计算机设备和存储介质
CN109308663A (zh) * 2018-06-22 2019-02-05 河海大学 一种基于水期权交易的水资源风险对冲方法
CN109523118A (zh) * 2018-10-11 2019-03-26 平安科技（深圳）有限公司 风险数据筛选方法、装置、计算机设备和存储介质
CN110033378A (zh) * 2019-01-29 2019-07-19 阿里巴巴集团控股有限公司 一种资源配置方法、装置及电子设备
CN110008986B (zh) * 2019-02-19 2023-05-12 创新先进技术有限公司 批量风险案件识别方法、装置及电子设备
CN110163407B (zh) * 2019-04-02 2023-01-17 创新先进技术有限公司 量化策略的优化方法及装置
CN110363647A (zh) * 2019-06-21 2019-10-22 深圳壹账通智能科技有限公司 风险指数的评估方法和装置
CN110442712B (zh) * 2019-07-05 2023-08-22 创新先进技术有限公司 风险的确定方法、装置、服务器和文本审理系统
CN110458601B (zh) * 2019-07-09 2023-11-24 平安科技（深圳）有限公司 资源数据的处理方法、装置、计算机设备和存储介质
CN110675248A (zh) * 2019-09-24 2020-01-10 支付宝(杭州)信息技术有限公司 一种账户评估方法、装置及设备

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103297267A (zh) * 2013-05-10 2013-09-11 河北远东通信系统工程有限公司 一种网络行为的风险评估方法和系统
CN106779723A (zh) * 2016-12-26 2017-05-31 中国银联股份有限公司 一种移动终端风险评估方法及装置
WO2019222017A1 (en) * 2018-05-18 2019-11-21 Beckman Coulter, Inc. Test panel analysis
CN109598414A (zh) * 2018-11-13 2019-04-09 阿里巴巴集团控股有限公司 风险评估模型训练、风险评估方法、装置及电子设备
CN111310784A (zh) * 2020-01-14 2020-06-19 支付宝(杭州)信息技术有限公司 资源数据的处理方法及装置

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
YANG YANG;WANG FENG-BIN;DAI PENG-JIE: "The Relationship between Institutional Isomorphism and Decision-making Rights Allocation in Large Group Companies", CHINA INDUSTRIAL ECONOMICS, 17 January 2016 (2016-01-17), pages 114 - 129, XP055830023, DOI: 10.19581/j.cnki.ciejournal.2016.01.008 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113706336A (zh) * 2021-09-01 2021-11-26 知能汇融(北京)咨询有限公司 风险评估方法、装置、计算机设备和存储介质
CN113706336B (zh) * 2021-09-01 2024-02-02 知能汇融(北京)咨询有限公司 风险评估方法、装置、计算机设备和存储介质

Also Published As

Publication number Publication date
CN113780404A (zh) 2021-12-10
CN111310784B (zh) 2021-07-20
CN113780404B (zh) 2024-11-26
CN111310784A (zh) 2020-06-19

Similar Documents

Publication Publication Date Title
WO2021143370A1 (zh) 资源数据的处理方法及装置
US9256693B2 (en) Recommendation system with metric transformation
TW201928848A (zh) 基於圖結構模型的信用風險控制方法、裝置以及設備
CN108446817B (zh) 确定业务对应的决策策略的方法、装置和电子设备
TW201935387A (zh) 反洗錢方法、裝置及設備
TW201928841A (zh) 訓練風控模型和風控的方法、裝置及設備
KR20160121806A (ko) 일시적 거래 한도 결정
TWI759539B (zh) 電腦設備及風控決策臨界值的確定裝置
TW201833851A (zh) 風控事件自動處理方法及裝置
CN111639687A (zh) 一种模型训练以及异常账号识别方法及装置
CN111539811A (zh) 风险账户的识别方法及装置
CN112506935A (zh) 数据处理方法、装置、电子设备、存储介质及程序产品
WO2019019346A1 (zh) 资产配置策略获取方法、装置、计算机设备和存储介质
US20230393813A1 (en) Sorting apparatus and method
Akyüz et al. Location and allocation based branch and bound algorithms for the capacitated multi-facility Weber problem
CN113792170A (zh) 图数据划分方法、装置和计算机设备
WO2022017167A1 (zh) 一种信息处理方法、系统、电子设备及存储介质
CN114611850A (zh) 业务分析方法、装置及电子设备
CN114581223B (zh) 分配任务处理方法、设备、分布式计算系统及存储介质
CN106886674A (zh) 一种地理位置距离批量计算方法及装置
WO2023093070A1 (zh) 面向智能城市网络资源的相关性分析方法及装置
CN118446744B (zh) 基于人工智能的权益数据转换方法及系统
CN115545364A (zh) 节点评价方法、装置、电子设备及可读存储介质
WO2021239084A1 (zh) 基于区块链的分佣
Madjoukeng et al. Fast k-means with stable instance sets

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20914671

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20914671

Country of ref document: EP

Kind code of ref document: A1